If I'm reading the current status graphs correctly they're regularly seeing 1.5-2.5 second queries and 400ms direct fetches. That seems unbearably bad. Can anyone running mission critical apps on the platform comment?
We're in the process of doing an emergency move over to EC2 because of these issues. That being said, I must admit that this is a really decent response on Google's part. Unfortunately the problems go deeper than just these latest problems with datastore.
We're thinking of porting over to EC2. If anyone knows of any reliable tool that supports this for production, please feel free to shout. Otherwise, we'll probably be building something ourselves...
If you write your own, would you consider open sourcing it?
Re: open source. We might just brute force it for now. If not, we'll definitely pretty it up at some point and open source it. At the very least we can publish our learnings.